TrotterNet has moved to its own server, at http://trotternet.com. Come check it out.
Well, I’m going to take another shot at actually writing this blog. When I first started this enterprise last year, I was appalled at the quality of the HTML that was emitted by Radio Userland. Its the very worst ‘tag soup’ that you can imagine. Not only that, but when it uses CSS, it does so in the worst possible way, by including the entire set of style definitions at the top of each and every file that it published. This means that not only do you lose the benefit of reduced file size that CSS normally provides, but you also lose the ability to modify a style definition for the entire site in a single location. Yech!
Since I’m a big advocate of CSS and standards-based markup, the thought of having my own site published in this manner was both embarrassing and frustrating. On the other hand, I’m an old-time Frontier programmer, so I thought that I could just delve into the object database, and modify the publishing macros so that they emitted proper, modern markup. Unfortunately, just as I was about to embark upon that little project, things got really, really busy at work, so I had to shelve the idea. And rather than endure the embarrassment of having my blog published using old school markup, I decided to just shelve it as well.
So, now things are a little less hectic at work, and I’m going to take another look at the publication macros. I haven’t started that project yet, but I’m going to start writing anyway. Hopefully, by the time anyone actually reads these words, I’ll have gotten the markup to the point where I won’t have to apologize for it.
That’s the plan anyway. We’ll see how it goes…
I know that nobody is reading this right now, so this is addressed to posterity. I haven’t posted in about 6 weeks because things at work have gotten WAY busy, and it’s likely to continue that way for a while longer. I hope to get back into the flow of things soon, but you never know. But since nobody’s reading anyway, it doesn’t really matter now, does it?
Here is a copy of the rather lengthy comment that I posted on Ben Hammersley’s weblog concerning a discussion about a new news aggregator product called NewsMonster:
I haven’t tried NewsMonster yet, but based on the discussion, it appears that the functionality that it most closely resembles is the “Offline Web Pages” feature of Internet Explorer for Windows. It also would appear that most people contributing to this discussion have not used this feature before, and therefore don’t appreciate just how valuable it is. If you haven’t used it, here’s a quick overview:
Offline Web Pages drives Internet Explorer just as if a live user were driving it. It stores complete web pages and all linked images and other content elements in IE’s regular cache. Its completely user configurable: it can store complete sites or just single pages depending on the URL; it can recursively dive down up to 3 (I think) levels deep; it can follow links to “outside” sites or stay within the domain specified by the initial URL; it can run on a schedule, on system events like startup or shutdown, or on demand; it can traverse and cache a single site, or a whole list of sites.
From the user’s perspective, you just run IE, put it into offline mode, then browse the site(s) as you would normally. There’s no difference between that and browsing the site online, except that the offline experience is blazingly fast, much faster than browsing online even over DSL or other broadband.
The way I used to use this feature was as follows: I have a half-hour train ride to and from work every day. I had my laptop set to download a list of sites every weekday morning at 5 a.m. and again in the afternoon at 4 p.m. The sites included CNET, NYT-Tech, Wired, GMSV and a few others. I could then read the news on the train using my laptop with IE in offline mode. This was a tremendous time-saver for me. I’ve since switched to using a Pocket PC for the train ride, but I still use Offline Web Pages for a few sites that I look at in the evenings at home.
Remember that the vast majority of web users still are stuck with 56K dialup, and will be for years to come. Using Offine Web Pages vastly improves the experience of browsing the web in that environment, as well as extending the availability of the web into situations where it isn’t currently accessable. Are Offline Web Pages inefficient from a server perspective? Certainly. Nevertheless, the feature is invaluable under certain circumstances.
KEY POINT: If Offline Web Pages obeyed the Robot Exclusion Protocol, it would render this valuable feature completely useless.
So what the answer? First, is to recognize that IE’s Offline Web Pages and (apparently) NewsMonster are neither robots in the “classic” sense of search engines, nor are they flesh-and-blood users, but are a hybrid of the two. The solution should be twofold:
First, the offline user agents need to be very smart and efficient. They shouldn’t try and download content that they already have in their cache. (Sites like CNET which have multiple CMS-generated URLs that point to the same article complicate this.) And they should try and learn from the user’s history and only download pages that the user is likely to actually read–which is easier said than done!
Second, the Robot Exclusion Protocol is ancient by Internet standards, and could probably use an update to better handle this situation. Perhaps it could redirect bots to an alternate URL which would allow them operate more efficiently. Or maybe there’s already some other technology which would be more appropriate.
Dave says: Last year on this day I asked if tables are really evil. Now one year later, I long for the simplicity of tables, … If it ain’t broke, dont fix it.
The trouble is, it is broke–it being the late ’90s set of HTML kludges and tricksneeded to get the pre-standards-compliant browsers to display a particular layout. The problems with these methods have been extensively documented elsewhere, but Dave just doesn’t see the brokeness as being what it is. Oh well…
CSS is hard. I went through the learning curve myself, and I’ve shepherded several of my designers and developers throught it as well, and it’s hard. I think the reason CSS is hard is because it’s totally unlike anything that most developers have ever encountered before. It’s not like learning a new programming language once you’ve already got one under your belt. There’s a philosophical hump that you have to get over: totally separating formatting from structure is a fundamentally different way of constructing web pages. But once you’ve gotten over the hump and experienced the benefits of the new approach, you won’t go back to the old way unless you’re forced to.
So even though CSS is hard to learn, and has browser support issues, in my estimation it’s still much better than the old approach. In general, I’m more concerned with forward compatibility than I am with backward compatibility. I’m more concerned that my content will usable on next year’s SmartPhones than I am about IE 3. If I have to learn something new, that’s okay because constant re-education is an integral part of the profession that I chose 28 years ago. The browser problems will fade over time (to be replaced by news ones, no doubt), and the learning curve will flatten out as better books and tutorials are published and the evolution of the development tools continues.
Speaking of development tools, if you’re trying to learn CSS or developing sites using CSS, you really need to be using TopStyle by Bradbury Software. Its the best thing out there right now for editing CSS, and its very reasonably priced.
Software Success Has India Worried. Is the United States going to start turning its back on outsourcing, the lifeblood of India’s software and services industry? By Saritha Rai.
Remember the hue and cry over expanding the H1B visa program a few years back? The IT outsourcing industry in India (and other countries) is a direct result of H1B visas. What happens is that highly educated Indians come to the U.S. and work for tech companies for a few years, gaining valuable experience and exposure to American culture and business practices. They then return to India and go to work for these outsourcing companies, which in turn compete for IT service contracts, under-bidding U.S. companies due to their lower wage expenses. Now, while some would use this situation to argue against the H1B program or the practice of allowing foreign access to U.S. high tech education and jobs, I’ve got a somewhat different perspective. While reducing or eliminating the H1B visas might be a good idea, what we really should be doing is aggressively recruiting these people to immigrate to the United States permanently and become U.S. citizens. I’d rather have them here, working for American companies in America, paying U.S. taxes, etc., than have the jobs they represent exported to Asia.
TiVo criticised for “invading privacy.” TiVo, the tapeless TV recording system, causes a row in the US after it automatically records the Discovery Channel.
I really, really want to own one of these devices, but this is exactly the kind of B.S. that keeps my wallet closed. I will fully control any device that is in my home, or I won’t buy it. In this particular case, I don’t so much mind the fact that it records Discovery Channel as long as it doesn’t take up any space for the things that I do want recorded. What really steams me is that it leaves the channel set to DC so that when you turn your TV on the next day, guess what you see?
I just re-posted my first post from a few days ago. Not content to simply install Radio and just use it, I managed to delete my first post this morning. Quite simple to do, actually. I installed Radio on a second computer, and when it upstreamed, it overwrote my initial post with a blank one. I recovered the text from the rss.xml file, and pasted it back and re-posted. Everything’s back to the way it was except for the date. [Update: got it back on 2/7 where it belongs.]
If I had been willing to wait until I got home tonight, I could have re-posted from my first Radio install, but this was more of a challenge. I learned a bit about where Radio stores it’s data, and I was a little surprised about how it’s done. Most of the configuration info is in the www folder, but all of the actual content of posts are contained in the .root files in the Data Files folder. And while you can change the location of the www folder, theres no obvious way to change the location of the other folders–at least, not that I’ve found yet.
Why does this matter to me? Because I work from two different locations, home and work, and sometimes from other places as well. I used to haul a laptop back and forth every day, but I gradually migrated to having two desktop systems with the laptop being used mainly as a portable hard drive. I commute by train so the weight of the laptop is a primary concern. I had an ultralight subnotebook (under 3-lbs), but it wasn’t nearly powerful enough to be used as my primary system. Then, when my 5-year-old son spilled a drink on the laptop last summer, I decided to wait on replacing it, and got a portable USB 2.0 30GB hard drive instead.
So, I’ve got a nearly identical set of applications installed on both home and work desktops, and I use the portable hard drive to store the data files. If anyone is interested in exactly how I got things setup, let me know and I’ll post the details. Suffice it to say that it’s important to be able to have two different installs on two different machines share the same common set of data files. Concurrency is not an issue since it’s only me using it. Radio makes this setup difficult because it stores it’s files–system and user–all mixed together within it’s Program Files subdirectory. And while it allows me to change the location of the www directory, there isn’t a user-visible preference to change the location of some of the other critical files. I’m sure there’s some path variable buried somewhere in the object database that controls this, but until I find where it is, I’ve got another idea in mind that will solve my problem. We’ll see…
I’m a programmer by training, so I had to start that way. My name is Scott Trotter, and I’m the Executive Producer and Technical Director of Paris France, an award-winning interactive media (a.k.a. web site) design shop located in Portland, Oregon. We use titles like that to emphasize to our clients that we’re a media production company rather than a technology company, but anywhere else I’d be the VP of Engineering or a Program Manager.
I’ve been meaning to start a weblog for quite a long time, but I could never seem to find the time to do it. Everytime there’s a flurry of posts on things like “what’s wrong with Flash,” or “CSS is a black hole,” or “what to do about the space program,” I’ve been itching to put my 2-cents in, but I’ve always been too busy to actually do it. I’m still too busy, but I’m going to make the time to write about the issues that are important to me.
I’m an original Mac owner (Jan 30, 1984), and I bought a copy of ThinkTank when it came out, so I’ve known who Dave Winer (and his brother Peter) were for a long time. I was a beta tester for Frontier 1.0, and have been reading DaveNet (and later Scripting News) since the Wired days, so when it came time to choose a blogging tool, there was only one choice: Radio Userland.
This weblog wont be like Dave’s or a lot of others, with dozens of links and brief comments every day. I’ll be lucky if I can find the time to post something 3 or 4 times a week, so I’ll only do so if I think I can make a worthwhile contribution to whatever topic is the subject of discussion. I’ll probably be playing around with the design, and eventually moving the site to my own server, both of which I understand are possible with Radio, so stay tuned.